i have a router with 8000k interfaces (because of isdn-controller).
and from time to time cacti is recaching the routers interfaces because the returned values are nonsense or U.
02/20/2008 01:29 PM
Code: Select all
02/20/2008 01:29:03 PM - SPINE: Poller[0] Host[35] DS[748] SNMP: v2: ROUTER, dsname: drops_in2, oid: .1.3.6.1.4.1.9.2.2.1.1.26.1, value: 153
02/20/2008 01:29:03 PM - SPINE: Poller[0] Host[35] DS[748] SNMP: v2: ROUTER, dsname: drops_out2, oid: .1.3.6.1.4.1.9.2.2.1.1.27.1, value: 0
02/20/2008 01:29:03 PM - SPINE: Poller[0] Host[35] DS[749] SNMP: v2: ROUTER, dsname: drops_in2, oid: .1.3.6.1.4.1.9.2.2.1.1.26.2, value: 5
02/20/2008 01:29:03 PM - SPINE: Poller[0] Host[35] DS[749] SNMP: v2: ROUTER, dsname: drops_out2, oid: .1.3.6.1.4.1.9.2.2.1.1.27.2, value: 0
Code: Select all
02/20/2008 01:30:02 PM - SPINE: Poller[0] Host[35] DS[748] SNMP: v2: ROUTER, dsname: drops_in2, oid: .1.3.6.1.4.1.9.2.2.1.1.26.1, value: 0
02/20/2008 01:30:02 PM - SPINE: Poller[0] Host[35] DS[748] SNMP: v2: ROUTER, dsname: drops_out2, oid: .1.3.6.1.4.1.9.2.2.1.1.27.1, value: 0
02/20/2008 01:30:02 PM - SPINE: Poller[0] Host[35] DS[749] SNMP: v2: ROUTER, dsname: drops_in2, oid: .1.3.6.1.4.1.9.2.2.1.1.26.2, value: 0
02/20/2008 01:30:02 PM - SPINE: Poller[0] Host[35] DS[749] SNMP: v2: ROUTER, dsname: drops_out2, oid: .1.3.6.1.4.1.9.2.2.1.1.27.2, value: 0
02/20/2008 01:30:04 PM - PCOMMAND: Poller[0] Host[35] RECACHE: Re-cache successful.
02/20/2008 01:30:04 PM - CMDPHP: Poller[0] DEBUG: SQL Exec: "delete from poller_command where poller_id=0"
02/20/2008 01:30:04 PM - RECACHE STATS: RecacheTime:0.2754 HostsRecached:1
Code: Select all
02/20/2008 01:31:03 PM - SPINE: Poller[0] Host[35] DS[748] SNMP: v2: ROUTER, dsname: drops_in2, oid: .1.3.6.1.4.1.9.2.2.1.1.26.1, value: 153
02/20/2008 01:31:03 PM - SPINE: Poller[0] Host[35] DS[748] SNMP: v2: ROUTER, dsname: drops_out2, oid: .1.3.6.1.4.1.9.2.2.1.1.27.1, value: 0
02/20/2008 01:31:03 PM - SPINE: Poller[0] Host[35] DS[749] SNMP: v2: ROUTER, dsname: drops_in2, oid: .1.3.6.1.4.1.9.2.2.1.1.26.2, value: 5
02/20/2008 01:31:03 PM - SPINE: Poller[0] Host[35] DS[749] SNMP: v2: ROUTER, dsname: drops_out2, oid: .1.3.6.1.4.1.9.2.2.1.1.27.2, value: 0Code: Select all
02-20-08.13:30:04 element: ROUTER - Queue Drops - Fa0/0 alertstat: 0 graph_id: 750 thold_low: thold_hi: 100 rra: trigger: 1 triggerct: 0 current: -153 logset: 1
02-20-08.13:30:04 element: ROUTER - Queue Drops - Fa0/0 alertstat: 0 graph_id: 750 thold_low: thold_hi: 100 rra: trigger: 1 triggerct: 0 current: 0 logset: 1
02-20-08.13:30:04 element: ROUTER - Queue Drops - Fa0/1 alertstat: 0 graph_id: 751 thold_low: thold_hi: 100 rra: trigger: 1 triggerct: 0 current: -5 logset: 1
02-20-08.13:30:04 element: ROUTER - Queue Drops - Fa0/1 alertstat: 0 graph_id: 751 thold_low: thold_hi: 100 rra: trigger: 1 triggerct: 0 current: 0 logset: 1AND
ROUTER - Queue Drops - Fa0/0 went above threshold of 100 with 4294967448
I rebuilded poller-cache -> no change.
I reindexed the interfaces -> i tried "uptime goes backwards" - "index count changed" - "verify all fields" and "none" to avoid recache-events -> no change
I tried to switch from spine to cmd.php -> no change
Question: Why is cacti getting a "0" instead of the real value from the router? I checked the router and the snmp-counters. Everything is ok!
Is there any way to avoid that "recache-events"?
Any ideas how i can fix this issue?
i am using Cacti 0.8.7a + Thold 3.9 + PIA 1.4 + newest spine
but i will do an update this week...
best regards,
Markus
